public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jonathan Wakely <jwakely@redhat.com>
To: Jakub Jelinek <jakub@redhat.com>
Cc: libstdc++@gcc.gnu.org, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] libstdc++: Workaround kernel-headers on s390x-linux
Date: Mon, 22 Apr 2024 16:41:56 +0100	[thread overview]
Message-ID: <CACb0b4nUO7zRWL5ycwD_+mD_G6M0c4zzxp5wYsZTWwOMK5M5-w@mail.gmail.com> (raw)
In-Reply-To: <ZiaEN0Vw5yPRXZYF@tucnak>

On Mon, 22 Apr 2024 at 16:37, Jakub Jelinek <jakub@redhat.com> wrote:
>
> Hi!
>
> We see
> FAIL: 17_intro/headers/c++1998/all_attributes.cc   (test for excess errors)
> FAIL: 17_intro/headers/c++2011/all_attributes.cc   (test for excess errors)
> FAIL: 17_intro/headers/c++2014/all_attributes.cc   (test for excess errors)
> FAIL: 17_intro/headers/c++2017/all_attributes.cc   (test for excess errors)
> FAIL: 17_intro/headers/c++2020/all_attributes.cc   (test for excess errors)
> FAIL: 17_intro/names.cc  -std=gnu++17 (test for excess errors)
> on s390x-linux.
> The first 5 are due to kernel-headers not using uglified attribute names,
> where <asm/types.h> contains
> __attribute__((packed, aligned(4)))
> I've filed a downstream bugreport for this in
> https://bugzilla.redhat.com/show_bug.cgi?id=2276084
> (not really sure where to report kernel-headers issues upstream), while the
> last one is due to <sys/ucontext.h> from glibc containing:
>   #ifdef __USE_MISC
>   # define __ctx(fld) fld
>   #else
>   # define __ctx(fld) __ ## fld
>   #endif
>   ...
>   typedef union
>     {
>       double  __ctx(d);
>       float   __ctx(f);
>     } fpreg_t;
> and g++ predefining -D_GNU_SOURCE which implies define __USE_MISC.
>
> The following patch adds a workaround for this on the libstdc++ testsuite
> side, ok for trunk?

OK, thanks.


>
> 2024-04-22  Jakub Jelinek  <jakub@redhat.com>
>
>         * testsuite/17_intro/names.cc (d, f): Undefine on s390*-linux*.
>         * testsuite/17_intro/headers/c++1998/all_attributes.cc (packed): Don't
>         define on s390.
>         * testsuite/17_intro/headers/c++2011/all_attributes.cc (packed):
>         Likewise.
>         * testsuite/17_intro/headers/c++2014/all_attributes.cc (packed):
>         Likewise.
>         * testsuite/17_intro/headers/c++2017/all_attributes.cc (packed):
>         Likewise.
>         * testsuite/17_intro/headers/c++2020/all_attributes.cc (packed):
>         Likewise.
>
> --- libstdc++-v3/testsuite/17_intro/names.cc.jj 2024-02-02 22:13:29.575359165 +0100
> +++ libstdc++-v3/testsuite/17_intro/names.cc    2024-04-22 16:54:35.696710752 +0200
> @@ -270,6 +270,12 @@
>  #undef u
>  #endif
>
> +#if defined (__linux__) && defined (__s390__)
> +// <sys/ucontext.h> defines fpreg_t::d and fpreg_t::f
> +#undef d
> +#undef f
> +#endif
> +
>  #if defined (__linux__) && defined (__sparc__)
>  #undef y
>  #endif
> --- libstdc++-v3/testsuite/17_intro/headers/c++1998/all_attributes.cc.jj        2024-01-03 12:08:09.603789437 +0100
> +++ libstdc++-v3/testsuite/17_intro/headers/c++1998/all_attributes.cc   2024-04-22 17:11:04.734990469 +0200
> @@ -29,7 +29,11 @@
>  # define noreturn 1
>  # define visibility 1
>  #endif
> +#ifndef __s390__
> +// kernel-headers <asm/types.h> uses __attribute__((packed,aligned(4))) on
> +// S390.
>  #define packed 1
> +#endif
>  #define pure 1
>  // glibc's sysdeps/unix/sysv/linux/arm/sys/ucontext.h uses this on ARM.
>  #ifndef __arm__
> --- libstdc++-v3/testsuite/17_intro/headers/c++2011/all_attributes.cc.jj        2024-04-22 17:11:20.028805205 +0200
> +++ libstdc++-v3/testsuite/17_intro/headers/c++2011/all_attributes.cc   2024-04-22 17:12:01.531302456 +0200
> @@ -29,7 +29,11 @@
>  # define visibility 1
>  #endif
>  #define no_unique_address 1
> +#ifndef __s390__
> +// kernel-headers <asm/types.h> uses __attribute__((packed,aligned(4))) on
> +// S390.
>  #define packed 1
> +#endif
>  #define pure 1
>  // glibc's sysdeps/unix/sysv/linux/arm/sys/ucontext.h uses this on ARM.
>  #ifndef __arm__
> --- libstdc++-v3/testsuite/17_intro/headers/c++2014/all_attributes.cc.jj        2024-01-03 12:08:09.556790095 +0100
> +++ libstdc++-v3/testsuite/17_intro/headers/c++2014/all_attributes.cc   2024-04-22 17:12:20.630071099 +0200
> @@ -29,7 +29,11 @@
>  # define visibility 1
>  #endif
>  #define no_unique_address 1
> +#ifndef __s390__
> +// kernel-headers <asm/types.h> uses __attribute__((packed,aligned(4))) on
> +// S390.
>  #define packed 1
> +#endif
>  #define pure 1
>  // glibc's sysdeps/unix/sysv/linux/arm/sys/ucontext.h uses this on ARM.
>  #ifndef __arm__
> --- libstdc++-v3/testsuite/17_intro/headers/c++2017/all_attributes.cc.jj        2024-01-03 12:08:09.506790795 +0100
> +++ libstdc++-v3/testsuite/17_intro/headers/c++2017/all_attributes.cc   2024-04-22 17:12:38.751851577 +0200
> @@ -28,7 +28,11 @@
>  # define visibility 1
>  #endif
>  #define no_unique_address 1
> +#ifndef __s390__
> +// kernel-headers <asm/types.h> uses __attribute__((packed,aligned(4))) on
> +// S390.
>  #define packed 1
> +#endif
>  #define pure 1
>  // glibc's sysdeps/unix/sysv/linux/arm/sys/ucontext.h uses this on ARM.
>  #ifndef __arm__
> --- libstdc++-v3/testsuite/17_intro/headers/c++2020/all_attributes.cc.jj        2024-01-03 12:08:09.502790851 +0100
> +++ libstdc++-v3/testsuite/17_intro/headers/c++2020/all_attributes.cc   2024-04-22 17:12:56.043642110 +0200
> @@ -27,7 +27,11 @@
>  # define cold 1
>  # define visibility 1
>  #endif
> +#ifndef __s390__
> +// kernel-headers <asm/types.h> uses __attribute__((packed,aligned(4))) on
> +// S390.
>  #define packed 1
> +#endif
>  #define pure 1
>  // glibc's sysdeps/unix/sysv/linux/arm/sys/ucontext.h uses this on ARM.
>  #ifndef __arm__
>
>         Jakub
>


      reply	other threads:[~2024-04-22 15:42 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-22 15:37 Jakub Jelinek
2024-04-22 15:41 ` Jonathan Wakely [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CACb0b4nUO7zRWL5ycwD_+mD_G6M0c4zzxp5wYsZTWwOMK5M5-w@mail.gmail.com \
    --to=jwakely@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jakub@redhat.com \
    --cc=libstdc++@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).